Stephen Marcus McAleer

Tree Search for Language Model Agents

Illusory Attacks: Information-theoretic detectability matters in adversarial attacks

Confronting Reward Model Overoptimization with Constrained RLHF

Llemma: An Open Language Model for Mathematics

Toward Optimal Policy Population Growth in Two-Player Zero-Sum Games

AgentKit: Structured LLM Reasoning with Dynamic Graphs

Steering No-Regret Learners to Optimal Equilibria

Game-Theoretic Robust Reinforcement Learning Handles Temporally-Coupled Perturbations